docs(DeepSeek-V4): mark b200|small|pd-disagg + h200|small|{cp,pd-disagg} verified (#23689)

2026-04-25 11:57:25 +08:00
parent fd401c2fb4
commit d2c61acf25
2 changed files with 22 additions and 1 deletions
@@ -145,7 +145,14 @@ The generator currently picks values on the **conservative** side (mirroring an

 **Hopper (H200) note**

-The H200 image and checkpoint are currently being uploaded — public path coming shortly.
+H200 image (`lmsysorg/sglang:deepseek-v4-hopper`) and FP8 checkpoints
+(`sgl-project/DeepSeek-V4-Flash-FP8`, `sgl-project/DeepSeek-V4-Pro-FP8`) are
+publicly available.
+
+PD-Disagg recipes on H200 may require `docker run --privileged --ulimit memlock=-1`
+(or `--device /dev/infiniband:/dev/infiniband --cap-add IPC_LOCK`) so mooncake
+can discover the IB HCAs; without IB exposure mooncake silently falls back to
+TCP, which can lead to garbled KV transfer on large checkpoints.

 **Base model usage**